---
output:
  pdf_document: default
  html_document: default
---
# Overview
This file describes the replication material for "Audit experiments of racial discrimination and the importance of symmetry in exposure to cues" by Thomas Leavitt and Viviana Rivera-Burgos in *Political Analysis*.

## Computational Requirements

We ran the code on a 2023 MacBook Pro with an Apple M3 Max chip and 36 GB memory. The operating system is Sonoma 14.3.1. The R version is 4.3.2.

The R packages are ggplot2 (version 3.4.4), dplyr (version 1.1.4) and randomizr (version 1.0.0).

## Data

Files: "data_construction.R", "example_data.tab" and "example_data.RData"

Description: The file "data_construction.R" creates the simulated data in "example_data.tab" and "example_data.RData". The variable "z" is the binary treatment and the variable "y" is the binary outcome. The running time of "data_construction.R" is 0.02 seconds.

## Code and Output

Files: "install_packages.R", "fig_2_code.R", "supplement_examples.R", "master.R" and "CI_plot.pdf".

Description: The file "install_packages.R" installs ggplot2 (version 3.4.4), dplyr (version 1.1.4) and randomizr (version = 1.0.0). The running time of "install_packages.R" is 2.03 seconds. The file "fig_2_code.R" constructs 95 percent confidence intervals over the feasible space of the proportion of Always-Openers, eventually yielding Figure 2 in the manuscript. The running time of "fig_2_code.R" is 0.13 seconds. The file "supplement_examples.R" conducts analyses on two simple worked examples in the online supplement. The running time of "supplement_examples.R" is 0.04 seconds. The "master.R" file runs all of the .R files (including "install_packages.R") reflecting the sequence of the workflow. The total running time of "master.R" is 2.42 seconds. The file "CI_plot.pdf" is Figure 2 in the manuscript, generated by the code in "fig_2_code.R".